=Paper=
{{Paper
|id=Vol-3632/ISWC2023_paper_455
|storemode=property
|title=Forest Explorer: User-Friendly Navigation of Iberian Forestry Linked Open Data
|pdfUrl=https://ceur-ws.org/Vol-3632/ISWC2023_paper_455.pdf
|volume=Vol-3632
|authors=Jose M. Gimenez-Garcia,Guillermo Vega-Gorgojo,Cristobal Ordonez,Felipe Bravo
|dblpUrl=https://dblp.org/rec/conf/semweb/Gimenez-GarciaV23
}}
==Forest Explorer: User-Friendly Navigation of Iberian Forestry Linked Open Data==
https://ceur-ws.org/Vol-3632/ISWC2023_paper_455.pdf
Forest Explorer: User-Friendly Navigation of Iberian
Forestry Linked Open Data
José M. Giménez-García1 , Guillermo Vega-Gorgojo1,2 , Cristóbal Ordóñez2 and
Felipe Bravo2
1
Group of Intelligent and Cooperative Systems, Universidad de Valladolid, Spain
2
Instituto Universitario de Investigación en Gestión Forestal Sostenible, iuFOR, Universidad de Valladolid, Spain
Abstract
The Forest Explorer is a web application that allows easy browsing of the contents of the Forest Inventories
and Maps of the Iberian Peninsula. It accesses a linked open data source created from the original data
in the European Cross-Forest project. The application is available at https://forestexplorer.gsic.uva.es/
explorer/ and can be accessed with a simple web browser on desktops, tablets, and mobile phones. The
user interface hides the complexity of the underlying technologies, providing an interactive map to
navigate to the area of interest and presenting the forest data in appropriate detail. The application is
suitable for professional as well as casual use by science communicators, data journalists or citizens.
Keywords
Forestry Data, Geospatial Data, Linked Open Data, Map Visualizations, User Interfaces
1. Introduction
Modern forestry science increasingly relies on the use of large datasets [1]. Among
these, national forest inventories and land cover maps are of particular importance.
These datasets are created and published as open data by national governments, to be
used by different agents, including forest stakeholders, operational foresters, data and
environmental journalists, or interested citizens. However, these datasets are difficult
to use, especially if integrated access is required, as they are created following different
methodologies and published in a diversity of formats. Semantic Web technologies can
potentially facilitate the integration and promote the usage of these datasets, but the
lack of knowledge by forestry agents in these technologies makes it necessary for tools to
fill this gap.
Cross-Forest1 was a European project working on the integration of forest inventories
and land cover maps data from Spain and Portugal with Semantic Web technologies. The
source datasets included: (1) The Spanish National Forest Inventory (Inventario Forestal
ISWC Posters and Demos. November 06–10, 2023, Athens, Greece
Envelope-Open jm.gimenez.garcia@gsic.uva.es (J. M. Giménez-García); guiveg@tel.uva.es (G. Vega-Gorgojo);
angelcristobal.ordonez@uva.es (C. Ordóñez); felipe.bravo@uva.es (F. Bravo)
Orcid 0000-0001-7015-7896 (J. M. Giménez-García); 0000-0002-6916-9213 (G. Vega-Gorgojo);
0000-0001-5354-3760 (C. Ordóñez); 0000-0001-7348-6695 (F. Bravo)
© 2023 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 Inter-
national (CC BY 4.0).
CEUR
Workshop
CEUR Workshop Proceedings (CEUR-WS.org)
Proceedings
http://ceur-ws.org
ISSN 1613-0073
1
https://crossforest.eu/
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings
Nacional, IFNes ) based on the use of a of a 1x1 kilometer grid, with a plot2 at each
point of the grid. (2) The Portuguese National Forest Inventory (Inventário Florestal
Nacional, IFNpt ) contains generalized dendrometric data grouped by NUTS3 levels. (3)
The Spanish Forest Map (Mapa Forestal Español, MFE), a land cover map focused on
forestry patches4 . Each patch includes information about land use and the dominant
tree species. And (4) the Portuguese Land Use Map (Carta de Ocupação do Solo, COS)
a land cover map of general use.
In the Cross-Forest project the ontologies are developed using a bottom-up methodology.
First, modeling the relevant concepts and relations in each source. Second, identifying and
describing concepts and relations shared among several sources (such as species, positions,
and measures). And third, generalizing related concepts within the same domain (i.e.,
forest inventories and land use). This leads to a modular ontology comprised by three sets
of modules: (1) bottom-level ontologies corresponding to the sources, (2) top-level domain
ontologies for forest inventories and land use, and (3) general cross-domain ontologies for
shares concepts and relations. The data is generated using first Open-Source tools5 to
convert the original sources into csv and GeoJson files, and then SPARQL-Generate [2]
to transform these files to RDF. All the results are publicly available in the GitHub
repository of the project6 , as well as Spanish7 , Portuguese8 , and European9 data portals.
The data can be queried through a SPARQL endpoint10 and all IRIs for the Spanish
datasets are dereferenceable. Ontologies and data are aligned with relevant external
sources, such as Wikidata [3], DBpedia [4] , and the NCBI Bioontology11 .
However, potential users lack the necessary knowledge in Semantic Web technologies to
properly exploit these results. Hence, we present the Forest Explorer, a web application
that allows to navigate the content of the published LOD datasets, and download their
data for a selected area.
2. The Forest Explorer
The Forest Explorer makes the Iberian forest inventories and land cover maps accessible
to potential users. It makes use of the LOD datasets published in the Cross-Forest
project. Its logical architecture can be seen in Figure 1. The Map generator is in charge
of displaying the view for the end user, employing a base map obtained from the Map
server, and listening to the requests made from the different Feature12 Managers to show
markers, polygons, popups, or tooltips on top of the map. The activation of different
2
Sample locations where dendrometric observations (such as tree density or wood volume) are taken
3
Nomenclature of Territorial Units for Statistics. See http://ec.europa.eu/eurostat/web/nuts/
4
Homogeneous areas of land delimited by polygons
5
MDB Tools (https://github.com/mdbtools/mdbtools) and Mapshaper (https://github.com/mbloch/mapshaper)
6
https://github.com/Cross-Forest/
7
https://datos.iepnb.es/def/sector-publico/medio-ambiente/
8
https://snig.dgterritorio.gov.pt/
9
http://www.europeandataportal.eu/
10
https://forestexplorer.gsic.uva.es/sparql/
11
https://obofoundry.org/ontology/ncbitaxon.html
12
Features in the Forest Explorer include provinces, patches, plots, and trees
feature managers depends on the zoom level. The Data manager handles all data requests
from the Feature managers. This component communicates with the SPARQL endpoints
using a REST API created with CRAFTS [5], configured to use the Cross-Forest and
DBpedia13 endpoints. Data requests include geospatial queries for obtaining the features
in a bounding box, as well as information requests about the details of a feature. Upon
receipt of a request, the Data manager first checks if the result is already available in the
Data cache. In case of a miss, the Data manager sends a call to the REST API, which in
turn sends one or more SPARQL queries to the endpoints.
The user interface allows zooming and panning intuitively with the mouse or a touch
screen, as well as through explicit zoom and positioning buttons. The data displayed
depends on the zoom level. Figure 2 illustrates these levels, starting with the northwestern-
most region of the Iberian peninsula and zooming in into the Geres-Xures Transboundary
Biosphere Reserve14 , on the border between the north of Portugal and Spain. The
furthest level, shown in Subfigure 2(a), presents aggregated data by NUTS3 regions
of Portugal and Spain. Zooming in shows three different layers with the data of the
land cover maps (MFE and COS), each one with increasing detail, in order to facilitate
visibility and reduce data transfer. Information about land use, area, density of species,
and (in the case of Spain) canopy cover15 of a patch can be displayed by hovering over
it. Subfigure 2(b) shows one of such levels. When the zoom level is close enough, the
plots of the IFNes are displayed, as seen in Subfigure 2(c). Hovering over a plot shows
its associated dendrometric measures. Finally, maximum zoom level on a plot reveals
the surveyed trees, their position, and the dendrometric measures taken for each one of
them. Subfigure 2(d) shows the content of a plot.
A form in the upper left corner provides several functionalities: (1) navigate to a
municipality (obtained via GeoNames16 ); (2) disable or enable the layer for administrative
regions (when the zoom is far enough); (3) show or hide the plots from the IFNes (when
the zoom is close enough); (4) download the data shown in the screen in JSON format;
and (5) highlight a number of taxa (species, genus, or family) using different colors,
using either common or scientific names. Images and descriptions for each taxon can
be displayed when selecting them—see Subfigure 2(b). In all subfigures, the oaks are
highlighted in cyan, and pines are highlighted in brown, creating different shades of
color when the two genera coexist in different proportions. This form is available in all
zoom levels, but can be collapsed using the “ – ” button. Subfigures 2(a,b) show the menu
displayed, while Subfigures 2(c,d) show the menu collapsed.
The Forest Explorer makes visually accessible the map of species composition of an
area and make easy the access to its data. This facilitates operational foresters and forest
stakeholders make silvicultural simulations for selecting managing alternatives needed
for precision forestry, and helps environmental journalists or interested citizens visualize
forest composition or discover areas with the presence of an species.
The Forest Explorer is developed in JavaScript, to facilitate its deployment as a Web
13
To provide images and multilingual descriptions of tree species
14
https://www.reservabiosferageresxures.eu
15
The proportion of the forest covered by the vertical projection of the tree crowns [6]
16
https://www.geonames.org/
Map generator Map server
Province Patch Plot Tree
manager manager manager manager
CRAFTS CrossForest
endpoint
JSON Forest Explorer
Data manager SPARQL
API DBpedia
endpoint
Data cache
Figure 1: The Forest Explorer Architecture
a b
c d
Figure 2: Snapshots of the Forest Explorer user interface. (a) Northwest Spanish and Portuguese NUTS3
regions. (b) Northwest Spanish and Portuguese land cover maps. (c) Land cover maps in the Geres-Xures
transboundary biosphere, with plots displayed for the Spanish area as circles on top of the patches. (d)
Plot on the Spanish side of the Geres-Xures transboundary biosphere.
application. It is portable and can be used on any device with a modern browser (it has
been successfully tested with the latest versions of Google Chrome and Mozilla Firefox
on various smartphones, tablets, and personal computers). It is publicly available at
https://forestexplorer.gsic.uva.es/explorer/. A video recording showcasing the functionalities
of the Forest Explorer interface can be found at: https://doi.org/10.6084/m9.figshare.24131916
3. Conclusions and Future Work
The Forest Explorer is a Web application to navigate, visualize, and selectively download
the contents of the Spanish and Portuguese national forest inventories and land cover
maps without needing to understand the underlying Semantic Web technologies of which
it makes use. This makes this important data available to different forestry agents and
lay users without requiring from them any knowledge of Semantic Web technologies.
The Cross-Forest dataset and the Forest Explorer are living projects. Our current
and future work includes adding the temporal dimension (by combining previous and
upcoming new versions of inventories and land cover maps) and incorporating spatial
planning data. We also envision extending the approach to other territories beyond the
Iberian peninsula.
Acknowledgments
This work has been partially funded by the Spanish Ministry of Science and Innovation
through LOD.For.Trees (TED2021-130667B-I00) and IMFLEX (PID2021-1262750B-C22)
projects and by the Junta de Castilla y León through CLU-2019-01 and CL-EI-2021-05
projects, co-financed by the European Regional Development Fund.
References
[1] H. Pretzsch, Forest dynamics, growth, and yield, volume 684, Springer, 2009.
[2] M. Lefrançois, A. Zimmermann, N. Bakerally, A SPARQL Extension for Generating
RDF from Heterogeneous Formats. In: Proceedings of the 14th Extended Semantic
Web Conference (ESWC), 2017, pp. 35–50.
[3] D. Vrandecic, M. Krötzsch, Wikidata: a free collaborative knowledgebase, in: Com-
mun. ACM 57.10 (2014), 78–85.
[4] J. Lehmann et al., DBpedia A large-scale, multilingual knowledge base extracted
from Wikipedia, in: Semantic Web Journal 6.2 (2015), 167–195.
[5] G. Vega-Gorgojo, CRAFTS: Configurable REST APIs for Triple Stores, in: IEEE
Access 10 (2022), 32426–32441.
[6] S. Jennings, N. Brown, D. Sheil, Assessing Forest Canopies and Understorey Illumi-
nation: Canopy Closure, Canopy Cover and Other Measures, in: Forestry 71 (1999),
59–73.